The loss of an unknown biodiversity: Spatial gaps in plant survey and conservation in a Brazilian hotspot of biodiversity

Abstract

Overall, biodiversity sampling is biased in different ways, such as taxa, geography, ecosystem, and accessibility. In the Brazilian Cerrado biome, a biodiversity hotspot, survey gaps have been identified across several taxa, which can hinder conservation effectiveness. Here, we address spatial gaps and sampling bias in plant surveys across Cerrado using an exhaustive literature search, and calculated data completeness and deficiency. We analyzed spatial sampling gaps in the Cerrado ecoregions, protected areas (PAs) and priority areas for conservation (PCs), and bias in relation to the distance to universities, PAs, PCs, and environmental heterogeneity. We also assessed the surveyed sites conservation status at fine and large spatial scales using multitemporal land cover maps (1985–2023), and identified the relationship between the number of species sampled, total surveyed area, and completeness. We found 1445 surveys encompassing 12,881.37 ha surveyed in Cerrado. We found low spatial completeness across all ecoregions. The central and southern Cerrado had higher number of surveys, while the northern and western had lower, < 20 survey sites, and are losing natural vegetation at both fine and large spatial scales. Surveyed sites are close to universities, PAs, and PCs, in areas with low environmental heterogeneity, and had <50 % overlap with PAs and PCs. The number of species increased with the total area surveyed and completeness, indicating that plant diversity is likely underestimated in most ecoregions. It is essential to increase the number of plant surveys in under surveyed ecoregions before remnants are lost due to agricultural expansion.